Gravity dam online monitoring method and device based on artificial neural network, and storage medium

Abstract

The invention discloses a gravity dam online monitoring method and device based on an artificial neural network, and a storage medium, and relates to the field of hydropower engineering. The gravity dam online monitoring method based on the artificial neural network comprises the steps that a first threshold value of a gravity dam monitoring index is determined through a fixed limit value method, and the gravity dam monitoring index is used for reflecting the safety state of a gravity dam; determining a second threshold value of the gravity dam monitoring index through the artificial neural network and the historical operation data of the gravity dam; and under the condition that the difference value between the first threshold value and the second threshold value is larger than a first preset value, the gravity dam is monitored through the second threshold value. By the adoption of the technical scheme, the problem that in the prior art, only a fixed limit value method is adopted to determine the threshold value of the gravity dam monitoring index, and consequently monitoring of the gravity dam is not accurate is solved.

Description

Technical Field

[0001] The present application relates to the field of hydropower engineering, and more specifically, to an online monitoring method and device for a gravity dam based on an artificial neural network, and a storage medium. Background Art

[0002] In the field of hydropower engineering, gravity dams are important hydraulic structures, and their operational safety is directly related to the normal operation of power stations and the safety of the surrounding environment. However, the performance and safety status of gravity dams change dynamically over time. Traditional gravity dam safety monitoring methods completely rely on the fixed limit method to determine the threshold of monitoring indicators. They lack consideration of such dynamic changes and are unable to adjust the threshold of monitoring indicators according to the actual operating data of the dam and environmental changes, resulting in inaccurate monitoring of gravity dams.

[0003] In view of the problem that in the related technology, only the fixed limit method is used to determine the threshold value of the gravity dam monitoring index, resulting in inaccurate monitoring of the gravity dam, no effective solution has been proposed so far.

[0004] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0005] The embodiments of the present application provide a method and device for online monitoring of a gravity dam based on an artificial neural network, a storage medium, and a computer program product, so as to at least solve the problem in the related art that only a fixed limit method is used to determine the threshold of the gravity dam monitoring indicator, resulting in inaccurate monitoring of the gravity dam.

[0006] According to one aspect of an embodiment of the present application, there is provided an online monitoring method for a gravity dam based on an artificial neural network, comprising: determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; determining a second threshold value of the gravity dam monitoring index by the artificial neural network and historical operating data of the gravity dam; and when the difference between the first threshold value and the second threshold value is greater than a first preset value, using the second threshold value to monitor the gravity dam.

[0007] In an exemplary embodiment, after determining the second threshold of the gravity dam monitoring indicator through the artificial neural network and the historical operating data of the gravity dam, the method further includes: when the difference between the first threshold and the second threshold is less than the first preset value, using the first threshold to monitor the gravity dam.

[0008] In an exemplary embodiment, the first threshold value of the gravity dam monitoring index is determined by a fixed limit method, including: determining the safety margin of each indicator in the gravity dam monitoring index based on the historical operation data of the gravity dam within a preset time, wherein the safety margin is the allowable variation range of each indicator; determining the sum of the standard value of each indicator and the safety margin as the first threshold value, wherein the standard value is the safe operation value of the gravity dam under the various indicators.

[0009] In an exemplary embodiment, after the second threshold is used to monitor the gravity dam, it includes: when some or all of the indicators monitored by the gravity dam are less than the second threshold, determining that the gravity dam is in normal operation; when some or all of the indicators monitored by the gravity dam are greater than the second threshold, activating the alarm device of the gravity dam.

[0010] In an exemplary embodiment, before determining the second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operation data of the gravity dam, the method also includes: when it is determined that there are missing values ​​in the historical operation data, determining whether the number of missing values ​​is less than a second preset value; when it is determined that the number of missing values ​​is less than the second preset value and the number of remaining data in the historical operation data is greater than a third preset value, deleting all historical operation data at the time corresponding to the missing value, wherein the remaining data is the data in the historical operation data other than the missing value.

[0011] In an exemplary embodiment, before determining the second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operating data of the gravity dam, the method also includes: in the case where there is one or more groups of duplicate data in the historical operating data, retaining the target data that appears for the first time in the one or more groups of duplicate data, and deleting other data in the one or more groups of duplicate data except the target data.

[0012] In an exemplary embodiment, the gravity dam monitoring index includes at least one of the following: a dam crest horizontal displacement index used to characterize the movement change of the dam crest of the gravity dam in the horizontal direction; a dam crest vertical displacement index used to characterize the movement change of the dam crest of the gravity dam in the vertical direction; a dam foundation uplift pressure index used to characterize the vertical upward pressure caused by the seepage water on the bottom of the dam foundation of the gravity dam; and a dam seepage volume index used to characterize the seepage volume of the gravity dam.

[0013] According to another aspect of an embodiment of the present application, there is also provided an online monitoring device for a gravity dam based on an artificial neural network, comprising: a first determination module, used to determine a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; a second determination module, used to determine a second threshold value of the gravity dam monitoring index by using the artificial neural network and the historical operation data of the gravity dam; and a monitoring module, used to monitor the gravity dam using the second threshold value when the difference between the first threshold value and the second threshold value is greater than a preset value.

[0014] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned online monitoring method of a gravity dam based on an artificial neural network when running.

[0015] According to another aspect of an embodiment of the present application, there is also provided an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-mentioned online monitoring method of a gravity dam based on an artificial neural network through the computer program.

[0016] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0017] Through the present application, the first threshold value of the gravity dam monitoring index is determined by the fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; the second threshold value of the gravity dam monitoring index is determined by the artificial neural network and the historical operation data of the gravity dam; when the difference between the first threshold value and the second threshold value is greater than the first preset value, the second threshold value is used to monitor the gravity dam. Thus, the problem of inaccurate monitoring of the gravity dam caused by only using the fixed limit method to determine the threshold value of the gravity dam monitoring index in the related art is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1 It is a hardware structure block diagram of a computer terminal of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0021] Figure 2 is a flow chart of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0022] Figure 3 It is a schematic diagram of a gravity dam monitoring indicator architecture of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0023] Figure 4 is a schematic diagram of threshold determination of a gravity dam online monitoring method based on an artificial neural network according to an embodiment of the present application;

[0024] Figure 5 It is a schematic diagram of a fusion judgment rule of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0025] Figure 6 It is a schematic diagram of a fusion evaluation rule based on a BP neural network in an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0026] Figure 7 It is a schematic diagram of a fusion evaluation rule based on an ELMAN neural network in an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application;

[0027] Figure 8 It is a structural block diagram of an online monitoring device for a gravity dam based on an artificial neural network according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or a similar computing device. Taking running on a computer terminal as an example, Figure 1 1 is a hardware structure block diagram of a computer terminal of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (Central Processing Unit, MCU) or a programmable logic device (Field Programmable Gate Array, FPGA)) and a memory 104 for storing data, wherein the above-mentioned computer terminal may also include a transmission device 106 and an input and output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0031] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the online monitoring method of gravity dam based on artificial neural network in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0032] A wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0033] In this embodiment, a method for online monitoring of a gravity dam based on an artificial neural network is provided, which is applied to the above-mentioned computer terminal. Figure 2 is a flow chart of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application, such as Figure 2 As shown, the process includes the following steps:

[0034] Step S202, determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam;

[0035] Step S204, determining a second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operation data of the gravity dam;

[0036] Step S206: When the difference between the first threshold and the second threshold is greater than a first preset value, the second threshold is used to monitor the gravity dam.

[0037] Through the above steps, the first threshold value of the gravity dam monitoring index is determined by the fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; the second threshold value of the gravity dam monitoring index is determined by the artificial neural network and the historical operation data of the gravity dam; when the difference between the first threshold value and the second threshold value is greater than the first preset value, the second threshold value is used to monitor the gravity dam. This solves the problem in the related art that only the fixed limit method is used to determine the threshold value of the gravity dam monitoring index, resulting in inaccurate monitoring of the gravity dam.

[0038] All historical operating data of various monitoring indicators are obtained from the gravity dam, including but not limited to horizontal displacement of the dam crest, vertical displacement of the dam crest, uplift pressure of the dam foundation, seepage of the dam, etc. Based on the safety standards and industry specifications of the design stage, combined with the operating data of the dam in the first year, the preliminary safety threshold of each monitoring indicator, i.e., the first threshold, is determined. All historical operating data of the first preset number are used as the training set of the artificial neural network, and the prediction accuracy of the artificial neural network is verified by all historical operating data of the second preset number. The artificial neural network with the highest prediction accuracy, i.e., the smallest prediction error, is selected to predict all historical operating data of the third preset number to determine the second threshold of the gravity dam monitoring indicator. When the difference between the first threshold and the second threshold is greater than the first preset value (e.g., 5%), the second threshold is used to monitor the gravity dam.

[0039] Optionally, taking the horizontal displacement of the dam top in the gravity dam monitoring indicator as an example, it is assumed that the first threshold of the horizontal displacement of the dam top is determined to be ±10cm based on the operating data of the gravity dam from January 1, 2020 to December 31, 2020. The operating data from January 1 to December 31, 2020 is used as the training set to train the BP neural network and the ELMAN neural network respectively, and the operating data from January 1 to March 31, 2021 is used as the validation set to determine the model with the smallest prediction error, assuming that the prediction error of the ELMAN neural network is the smallest. The ELMAN neural network is used to predict the historical operating data from April 1 to December 31, 2021. It is assumed that the range of the horizontal displacement of the dam top predicted by the ELMAN neural network is ±9cm (i.e., the second threshold). Compare the difference between the first threshold (±10cm) and the second threshold (±9cm), and calculate the relative change. (10-9) / 10=0.1, i.e. 10%, which is greater than the first preset value (assuming 5%), the monitoring system of the gravity dam automatically updates the monitoring threshold of the horizontal displacement of the dam top to ±9 cm, i.e. adopting the second threshold as the new monitoring standard.

[0040] In an exemplary embodiment, after determining the second threshold of the gravity dam monitoring indicator through the artificial neural network and the historical operating data of the gravity dam, the method further includes: when the difference between the first threshold and the second threshold is less than the first preset value, using the first threshold to monitor the gravity dam.

[0041] Optionally, taking the horizontal displacement of the dam top in the gravity dam monitoring indicator as an example, assuming that the first threshold is ±10 cm and the second threshold is ±9.8 cm, the relative change between the first threshold and the second threshold is determined to be (10-9.8) / 10=0.02, that is, 2%, which is less than the first preset value (assuming 5%), then the monitoring threshold of the horizontal displacement of the dam top is ±10 cm, that is, the first threshold is used as the monitoring standard for the horizontal displacement of the dam top.

[0042] In an exemplary embodiment, the first threshold value of the gravity dam monitoring index is determined by a fixed limit method, including: determining the safety margin of each indicator in the gravity dam monitoring index based on the historical operation data of the gravity dam within a preset time, wherein the safety margin is the allowable variation range of each indicator; determining the sum of the standard value of each indicator and the safety margin as the first threshold value, wherein the standard value is the safe operation value of the gravity dam under the various indicators.

[0043] Optionally, obtain the historical operating data of the gravity dam in the first year, including daily monitoring records of indicators such as the horizontal displacement of the dam crest, the vertical displacement of the dam crest, the dam foundation uplift pressure and the dam seepage. Analyze the historical operating data of the first year to find out the range of variation of each indicator under normal operating conditions. Assume that in the historical operating data of the first year, the maximum horizontal displacement of the dam crest is +6cm and the minimum horizontal displacement is -6cm, so the safety margin is ±6cm. The range of the vertical displacement of the dam crest is +3cm to -3cm, and the safety margin is set to ±3cm. The fluctuation range of the dam foundation uplift pressure is 10kPa to 20kPa, and the safety margin is set to ±5kPa. The range of variation of the dam seepage is 5m 3 / s to 15m 3 / s, the safety margin is set to ±5m 3 / s. The sum of the standard value and the safety margin is determined as the first threshold, that is: the first threshold of the horizontal displacement of the dam crest is 0cm+6cm=+6cm and 0cm-6cm=-6cm, so the first threshold is set to ±6cm. The first threshold of the vertical displacement of the dam crest is 0cm+3cm=+3cm and 0cm-3cm=-3cm, so the first threshold is ±3cm. The first threshold of the dam foundation uplift pressure is 15kPa (assuming the standard value is 15kPa) + 5kPa=20kPa and 15kPa-5kPa=10kPa, so the first threshold is set to 10kPa to 20kPa. The first threshold of the dam seepage is 10m 3 / s(Assuming the standard value is 10m 3 / s)+5m 3 / s=15m 3 / s and 10m 3 / s-5m 3 / s=5m 3 / s, so the first threshold is set to 5m 3 / s to 15m 3 / s.

[0044] In an exemplary embodiment, after the second threshold is used to monitor the gravity dam, it includes: when some or all of the indicators monitored by the gravity dam are less than the second threshold, determining that the gravity dam is in normal operation; when some or all of the indicators monitored by the gravity dam are greater than the second threshold, activating the alarm device of the gravity dam.

[0045] Optionally, taking the horizontal displacement of the dam crest as an example, the second threshold value obtained by analyzing historical data through artificial neural network is ±7cm. This means that the horizontal displacement range of the dam crest under normal operating conditions should be maintained between -7cm and +7cm. If the real-time monitoring data shows that the horizontal displacement of the dam crest is +5.5cm, which is less than the upper limit of the second threshold of +7cm, it can be determined that the dam is in normal operating condition in terms of this indicator. If the horizontal displacement of the dam crest suddenly shows +7.2cm, this exceeds the upper limit of the second threshold of +7cm. The monitoring system of the gravity dam will immediately activate an alarm, indicating that there is an abnormality on the dam crest and further investigation is required.

[0046] In an exemplary embodiment, before determining the second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operation data of the gravity dam, the method also includes: when it is determined that there are missing values ​​in the historical operation data, determining whether the number of missing values ​​is less than a second preset value; when it is determined that the number of missing values ​​is less than the second preset value and the number of remaining data in the historical operation data is greater than a third preset value, deleting all historical operation data at the time corresponding to the missing value, wherein the remaining data is the data in the historical operation data other than the missing value.

[0047] Optionally, obtain monitoring indicator data of the gravity dam, including horizontal displacement of the dam crest, vertical displacement of the dam crest, dam foundation uplift pressure, dam seepage, etc. The data is collected from the whole year of 2021, a total of 365 days of data. Preprocess the acquired historical operation data, including data cleaning, and check whether there are outliers or missing values ​​in the data. Assuming that the dam foundation uplift pressure data on February 14, 2021 is missing, set the second preset value to 5% of the total number of data days, that is, 365 days × 5% = 18.25 days, rounded to 18 days. Since the number of missing values ​​is less than 18 days, that is, less than the second preset value, this means that data missing will not seriously affect the overall quality of the data. Set the third preset value to 90% of the total number of data days, that is, 365 days × 90% = 328.5 days, rounded to 328 days. We have 365-1 = 364 days of complete data, which is much larger than 328 days and meets the data volume requirements. Since the number of days with missing data is less than the second preset value and the number of days with complete data is greater than the third preset value, according to the preset rules, all monitoring indicator data on February 14, 2021 are deleted to avoid interference of partial data missing on model training.

[0048] In an exemplary embodiment, before determining the second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operating data of the gravity dam, the method also includes: in the case where there is one or more groups of duplicate data in the historical operating data, retaining the target data that appears for the first time in the one or more groups of duplicate data, and deleting other data in the one or more groups of duplicate data except the target data.

[0049] Optionally, assume that in the acquired data for 2022, there are two sets of duplicate data for the dam base uplift pressure data on June 15, 2022, that is, the first set of duplicate data is 15.2 kPa repeatedly recorded at three time points: 11:00, 13:00 and 15:00, and the second set of duplicate data is 12.3 kPa repeatedly recorded at three time points: 10:00, 14:00 and 16:00. According to the above processing rules, only the first data record in each set of duplicate data (i.e., 15.2 kPa at 11:00 and 12.3 kPa at 10:00) will be retained, and the data records at 13:00 and 15:00 in the first set of duplicate data and the data records at 14:00 and 16:00 in the second set of duplicate data will be deleted to avoid interference of these duplicate data on model training and monitoring systems.

[0050] In an exemplary embodiment, the gravity dam monitoring index includes at least one of the following: a dam crest horizontal displacement index used to characterize the movement change of the dam crest of the gravity dam in the horizontal direction; a dam crest vertical displacement index used to characterize the movement change of the dam crest of the gravity dam in the vertical direction; a dam foundation uplift pressure index used to characterize the vertical upward pressure caused by the seepage water on the bottom of the dam foundation of the gravity dam; and a dam seepage volume index used to characterize the seepage volume of the gravity dam.

[0051] The horizontal displacement of the dam top is usually caused by factors such as temperature changes in the dam body, water level fluctuations, structural aging or foundation settlement. Monitoring this indicator can help evaluate the stability of the dam body, detect possible structural displacement in time, prevent excessive horizontal movement of the dam body, and avoid potential safety risks. The vertical displacement index of the dam top can reflect the deformation of the dam body under the action of water pressure and deadweight, as well as the response of the dam body to environmental changes. The dam foundation uplift pressure reflects the potential impact of water pressure on the stability of the dam foundation, thereby helping to determine whether the dam foundation is abnormal and whether there is a risk of sliding or instability. The dam seepage index is used to characterize the seepage of the gravity dam. Abnormal seepage may be a signal of problems in the dam body or dam foundation anti-seepage system, such as cracks, holes, etc. Monitoring the seepage can help detect and deal with leakage problems early and prevent water pressure from causing irreversible damage to the dam structure.

[0052] In order to better understand the process of the above-mentioned online monitoring method for gravity dams based on artificial neural networks, the above-mentioned online monitoring method for gravity dams based on artificial neural networks is described below in combination with optional embodiments, but it is not used to limit the technical solution of the embodiments of the present application.

[0053] Figure 3 is a schematic diagram of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application, such as Figure 3 As shown, specifically including the following:

[0054] In the online monitoring system for gravity dam safety, monitoring indicators are mainly used to continuously monitor the health status of the gravity dam, water diversion system and powerhouse, among which this embodiment will focus on the monitoring of the gravity dam body and other key parts. Specifically, the key monitoring items include the following aspects: horizontal and vertical displacements of the dam crest, uplift pressure of the dam foundation and seepage volume of the dam. Among them, the horizontal and vertical displacements of the dam crest directly reflect the displacement changes of the dam body in the horizontal and vertical directions, and are the core indicators for evaluating the overall stability and structural safety of the dam. The uplift pressure of the dam foundation is used to monitor the changes in the internal water pressure of the dam foundation, which is crucial to judging the stability of the dam foundation, because an abnormal increase in uplift pressure may indicate a decrease in the bearing capacity of the foundation. The seepage volume of the dam is used to monitor the seepage of the dam. Sudden changes in the seepage volume may indicate potential waterproofing problems in the dam body or dam foundation.

[0055] For the dam body itself, in addition to the above key monitoring items, it also includes: deflection deformation and deformation of the joint between the dam foundation and the bedrock. Among them, deflection deformation is used to monitor the elastic deformation of the dam under load, and the margin and predicted value of this deformation are determined by the fixed limit method, historical extreme value method and trend method. The deformation of the joint between the dam foundation and the bedrock is used to monitor the deformation of the contact surface between the dam foundation and the bedrock, which is very important for evaluating the bonding strength between the dam and the foundation.

[0056] For the monitoring of deflection deformation, the following strategy is used to determine the monitoring indicators. In the initial stage, the monitoring indicators are determined by the fixed limit method, the historical extreme value method and the trend method, and are reflected in the form of margin (i.e., the first threshold) and the predicted value (i.e., the second threshold). In the subsequent operation, if the difference between the first threshold and the second threshold exceeds 5%, it indicates that the current displacement change exceeds the expected safety range. At this time, the monitoring system will automatically adjust and use the second threshold as the new monitoring indicator to monitor and evaluate the safety status of the gravity dam in real time. On the contrary, if the difference between the first threshold and the second threshold is less than or equal to 5%, it means that the displacement change of the gravity dam is within an acceptable range. At this time, the first threshold will be used as the monitoring indicator to ensure the continuity and stability of the monitoring strategy.

[0057] Obviously, the embodiments described above are only part of the embodiments of the present application, not all of the embodiments. In order to better understand the above-mentioned online monitoring method of gravity dam based on artificial neural network, the above process is described below in combination with the embodiments, but it is not used to limit the technical solutions of the embodiments of the present application, specifically:

[0058] In an optional embodiment, in combination Figure 4 The online monitoring method of a gravity dam based on an artificial neural network of the present application is further described. Figure 4 As shown, including the following:

[0059] In the first year after the gravity dam is put into operation, the operation data of all key monitoring surfaces are continuously collected, including but not limited to dam top displacement, uplift pressure, seepage flow, temperature, etc. The collected data are cleaned, outliers are removed, missing values ​​are filled, and data are standardized. The fixed limit method is used to determine the preliminary safety monitoring threshold, i.e., the first threshold, for the operation data of the gravity dam in the first year after preprocessing. All historical operation data since the operation of the gravity dam, including the operation data in the first year and the following years, are collected and deeply analyzed by artificial neural network. All historical operation data of the first preset number are used as the training set of the artificial neural network, and the prediction accuracy of the artificial neural network is verified by all historical operation data of the second preset number. The artificial neural network with the highest prediction accuracy, i.e., the smallest prediction error, is selected to predict all historical operation data of the third preset number to determine the second threshold of the gravity dam monitoring index. The first threshold and the second threshold are compared. If the difference between the first threshold and the second threshold is greater than 5%, the second threshold is used as the new monitoring standard to reflect the latest safety requirements of the operation status of the gravity dam. On the contrary, if the difference is less than 5%, it indicates that the operation status of the gravity dam is basically consistent with the safety standard set in the design stage, so the first threshold is continued to be used for monitoring.

[0060] In order to better understand the process of the above-mentioned gravity dam online monitoring method based on artificial neural network, the following is combined with Figure 5 — Figure 7 The optional embodiment illustrates the fusion evaluation rules in the above-mentioned gravity dam online monitoring method based on artificial neural network, but is not used to limit the technical solution of the embodiment of the present application.

[0061] Figure 5 is a schematic diagram of a fusion evaluation rule of an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application, such as Figure 5 As shown, specifically including the following:

[0062] Step S502: Establish a measurement point and monitoring database, which contains all existing monitoring point information and historical monitoring data, providing a data basis for subsequent analysis and dynamic update.

[0063] Step S504: Regularly check whether there are any new damaged measuring points in the monitoring system. If any new damaged measuring points are found, it is necessary to immediately update the measuring points and the monitoring database, remove or mark these damaged measuring points, so as to avoid them affecting subsequent analysis.

[0064] Step S506: On the premise of confirming that there are no newly added damaged measuring points, the current health status of key monitoring items such as dam crest horizontal displacement, dam crest vertical displacement, and dam foundation uplift pressure are analyzed by comparing current data with historical data.

[0065] Step S508: Continue to conduct in-depth analysis, pay attention to the status of specific monitoring parts such as the dam crest, water diversion tunnel, pressure pipeline, surge tank, etc., understand the operation of each part of the dam in more detail, and identify possible local problems.

[0066] Step S510: Analyze the status of the dam body, water diversion system, powerhouse, etc. as a whole monitoring object, integrate the status of local monitoring parts, and form a comprehensive assessment of the entire gravity dam and its related facilities.

[0067] Step S512: Based on the analysis results of the monitoring items, monitoring parts and monitoring objects, a comprehensive judgment is made to determine the health status of the gravity dam. If all analysis indicators are within the safe range, it can be confirmed that the dam is in a healthy operating state. On the contrary, if any indicator exceeds the normal range, a detailed inspection and maintenance procedure needs to be initiated to ensure the safety of the dam.

[0068] Figure 6 is a schematic diagram of a fusion evaluation rule based on a BP neural network in an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application, such as Figure 6 As shown, specifically including the following:

[0069] In the online monitoring system of gravity dam based on BP neural network, historical data input is the basis of network learning and prediction. These historical data include key indicators of dam operation, such as horizontal displacement and vertical displacement of dam crest, uplift pressure of dam foundation, and seepage of dam.

[0070] The core of the BP neural network lies in its hidden layer, or the middle layer, which is composed of a large number of neurons and processes data through complex weight connections. The role of the hidden layer is to extract deep features from the input data and use nonlinear transformations to map the data from a high-dimensional space to another feature space that can better reflect the health status of the dam. Each neuron calculates a weighted sum based on the product of its weight and the input data, and then transforms it through an activation function (such as Sigmoid or ReLU) to pass the result to the next layer. This process includes forward propagation, which is the flow of data from the input layer to the output layer, and backpropagation, which is a weight adjustment mechanism that calculates the error between the output value and the actual dam state, and backpropagates the error information back to the network, using the gradient descent method to update the weights to minimize the prediction error, thereby optimizing the network's prediction performance.

[0071] During the training process, the BP neural network can not only learn the association pattern between the monitoring indicators and the dam health status in the historical monitoring data, but also dynamically generate or adjust the thresholds of the monitoring indicators. These thresholds change over time, reflecting the true dynamic characteristics of the dam's operating status, avoiding the limitations and errors brought by the traditional fixed threshold method. By minimizing the error between the predicted value and the actual dam status, the network learns the normal operating range of the monitoring indicators and the abnormal thresholds, so that the health status and potential risks of the dam can be more accurately identified.

[0072] Once the BP neural network training is completed, it can be deployed in the real-time monitoring system to automatically analyze new monitoring data, dynamically generate thresholds for monitoring indicators, and compare them with actual monitoring values. If all monitoring indicator values ​​are within the dynamic threshold range, the dam is determined to be in a safe operating state; if any indicator exceeds the set threshold range, the monitoring system will trigger an alarm so that the target object of maintaining the gravity dam can take timely measures to ensure the safety and stable operation of the dam.

[0073] Figure 7 is a schematic diagram of a fusion evaluation rule based on an ELMAN neural network in an online monitoring method for a gravity dam based on an artificial neural network according to an embodiment of the present application, such as Figure 7 As shown, specifically including the following:

[0074] For gravity dam monitoring, historical data input includes historical records of monitoring indicators such as dam crest horizontal displacement, dam crest vertical displacement, dam foundation uplift pressure, dam seepage, etc. In order to make the data suitable for the input requirements of the ELMAN network, preprocessing is required, including data cleaning, normalization, and possible feature engineering to ensure data quality and improve model training efficiency and prediction accuracy.

[0075] Unlike BP neural networks, the hidden layer of the ELMAN network contains feedback connections, which allows the network to have a certain memory ability when processing time series data, that is, the network not only processes the input at the current time point, but also remembers and uses the information at previous time points to affect the current output. This memory ability enables the ELMAN network to capture the time series change rules of monitoring indicators and provide a richer information background for the prediction of dam operation status.

[0076] The association layer of the ELMAN network is the key to its processing of time series data. This layer receives the output from the hidden layer and then feeds these outputs back to the hidden layer, allowing the network to take into account the historical state when processing the input of continuous time points, thereby more accurately understanding and predicting the changing trends of monitoring indicators. This interaction between the association layer and the hidden layer enhances the network's ability to process time dependencies in sequence data, and plays an irreplaceable role in predicting the health of gravity dams.

[0077] Finally, the output layer of the network dynamically generates new thresholds based on the comprehensive processing results of the hidden layer and the association layer. These thresholds represent the normal range and abnormal range of the monitoring indicators at a specific time point, and they will be continuously optimized as the network learning progresses to reflect the true state of the dam over time. By minimizing the error between the predicted value and the actual dam state, the ELMAN network learns the normal operating mode and abnormal threshold of the monitoring indicators, so as to provide more accurate assessment and early warning of the health status of the dam.

[0078] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0079] In this embodiment, a gravity dam online monitoring device based on artificial neural network is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0080] Figure 8 : is a structural block diagram of an online monitoring device for a gravity dam based on an artificial neural network according to an embodiment of the present application, the device comprising:

[0081] A first determination module 80, configured to determine a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam;

[0082] A second determination module 82, configured to determine a second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam;

[0083] The monitoring module 84 is configured to monitor the gravity dam using the second threshold value when the difference between the first threshold value and the second threshold value is greater than a preset value.

[0084] Through the above device, the first threshold value of the gravity dam monitoring index is determined by the fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; the second threshold value of the gravity dam monitoring index is determined by the artificial neural network and the historical operation data of the gravity dam; when the difference between the first threshold value and the second threshold value is greater than the first preset value, the second threshold value is used to monitor the gravity dam. This solves the problem in the related art that only the fixed limit method is used to determine the threshold value of the gravity dam monitoring index, resulting in inaccurate monitoring of the gravity dam.

[0085] In an exemplary embodiment, the second determination module 82 is further configured to use the first threshold to monitor the gravity dam when the difference between the first threshold and the second threshold is smaller than the first preset value.

[0086] In an exemplary embodiment, the first determination module 80 is also used to determine the safety margin of each indicator in the gravity dam monitoring indicators based on the historical operation data of the gravity dam within a preset time, wherein the safety margin is the allowable variation range of the various indicators; determine the sum of the standard value of the various indicators and the safety margin as the first threshold value, wherein the standard value is the safe operation value of the gravity dam under the various indicators.

[0087] In an exemplary embodiment, the monitoring module 84 is also used to determine that the gravity dam is in normal operation when some or all of the monitoring indicators of the gravity dam are less than the second threshold value; and to activate the alarm device of the gravity dam when some or all of the monitoring indicators of the gravity dam are greater than the second threshold value.

[0088] In an exemplary embodiment, the second determination module 82 is also used to determine whether the number of missing values ​​is less than a second preset value when it is determined that there are missing values ​​in the historical operation data; when it is determined that the number of missing values ​​is less than the second preset value and the number of remaining data in the historical operation data is greater than a third preset value, delete all historical operation data at the time corresponding to the missing value, wherein the remaining data is the data in the historical operation data other than the missing value.

[0089] In an exemplary embodiment, the second determination module 82 is also used to retain the target data that appears for the first time in one or more groups of duplicate data, and delete other data in the one or more groups of duplicate data except the target data, when there is one or more groups of duplicate data in the historical operation data.

[0090] In an exemplary embodiment, the first determination module 80 is also used for the gravity dam monitoring indicators, including at least one of the following: a dam top horizontal displacement indicator for characterizing the movement change of the dam top of the gravity dam in the horizontal direction; a dam top vertical displacement indicator for characterizing the movement change of the dam top of the gravity dam in the vertical direction; a dam foundation uplift pressure indicator for characterizing the vertical upward pressure caused by the seepage water on the bottom of the dam foundation of the gravity dam; and a dam seepage volume indicator for characterizing the seepage volume of the gravity dam.

[0091] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0092] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0093] S1, determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam;

[0094] S2, determining a second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam;

[0095] S3: When the difference between the first threshold and the second threshold is greater than a first preset value, the second threshold is used to monitor the gravity dam.

[0096] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0097] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0098] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0099] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0100] S1, determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam;

[0101] S2, determining a second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam;

[0102] S3: When the difference between the first threshold and the second threshold is greater than a first preset value, the second threshold is used to monitor the gravity dam.

[0103] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0104] An embodiment of the present application further provides a computer program product, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program product, and when the computer program is executed by a processor, the steps of the method described in each embodiment of the present application are implemented.

[0105] Optionally, in this embodiment, the above computer program may be configured to implement the following steps when executed by a processor:

[0106] S1, determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam;

[0107] S2, determining a second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam;

[0108] S3: When the difference between the first threshold and the second threshold is greater than a first preset value, the second threshold is used to monitor the gravity dam.

[0109] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0110] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0111] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An online monitoring method for a gravity dam based on an artificial neural network, characterized in that: include: Determining a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; Determining a second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam; When the difference between the first threshold and the second threshold is greater than a first preset value, the second threshold is used to monitor the gravity dam.

2. The method for online monitoring of a gravity dam based on an artificial neural network according to claim 1, characterized in that: After determining the second threshold of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam, the method further includes: When the difference between the first threshold and the second threshold is smaller than the first preset value, the first threshold is used to monitor the gravity dam.

3. The method for online monitoring of a gravity dam based on an artificial neural network according to claim 1, characterized in that: The first threshold of the gravity dam monitoring index is determined by the fixed limit method, including: The safety margin of each indicator in the gravity dam monitoring index is determined according to the historical operation data of the gravity dam within a preset time, wherein the safety margin is the allowable variation range of each indicator; the sum of the standard value of each indicator and the safety margin is determined as the first threshold value, wherein the standard value is the safe operation value of the gravity dam under the various indicators.

4. The method for online monitoring of a gravity dam based on an artificial neural network according to claim 1, characterized in that: After the second threshold is used to monitor the gravity dam, the method includes: When some or all of the indicators of the gravity dam monitoring indicators are less than the second threshold, determining that the gravity dam is in a normal operating state; When part or all of the monitoring indicators of the gravity dam are greater than the second threshold, the alarm device of the gravity dam is activated.

5. The method for online monitoring of a gravity dam based on an artificial neural network according to claim 1, characterized in that: Before determining the second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam, the method further includes: In the case of determining that there are missing values ​​in the historical operation data, determining whether the number of the missing values ​​is less than a second preset value; When it is determined that the number of missing values ​​is less than the second preset value and the number of remaining data in the historical operation data is greater than the third preset value, all historical operation data at the moment corresponding to the missing value are deleted, wherein the remaining data is the data in the historical operation data other than the missing value.

6. The method for online monitoring of a gravity dam based on an artificial neural network according to claim 1, characterized in that: Before determining the second threshold value of the gravity dam monitoring indicator by using the artificial neural network and the historical operation data of the gravity dam, the method further includes: In the case that one or more sets of duplicate data exist in the historical operation data, the target data that appears for the first time in the one or more sets of duplicate data is retained, and other data except the target data in the one or more sets of duplicate data is deleted.

7. The method for online monitoring of a gravity dam based on an artificial neural network according to any one of claims 1 to 6, characterized in that: The gravity dam monitoring indicators include at least one of the following: A dam crest horizontal displacement index used to characterize the movement change of the dam crest of the gravity dam in the horizontal direction; A dam crest vertical displacement index used to characterize the movement change of the dam crest of the gravity dam in the vertical direction; A dam foundation uplift pressure index used to characterize the vertical upward pressure caused by the seepage water on the bottom of the dam foundation of the gravity dam; A dam seepage index used to characterize the seepage volume of the gravity dam.

8. An online monitoring device for a gravity dam based on an artificial neural network, characterized in that: include: A first determination module, configured to determine a first threshold value of a gravity dam monitoring index by a fixed limit method, wherein the gravity dam monitoring index is used to reflect the safety status of the gravity dam; A second determination module is used to determine a second threshold value of the gravity dam monitoring indicator through the artificial neural network and the historical operation data of the gravity dam; A monitoring module is used to monitor the gravity dam using the second threshold value when the difference between the first threshold value and the second threshold value is greater than a preset value.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 7 when executed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images